Preliminary Comments on
Survey of Local Regulatory Practices and Manufactured Homes (200612-2528-003)
*** Responses are in red text.
You’ve indicated that states and metropolitan localities will be provided the results of the study. Please clarify the type of information that they will be provided? For example, do you plan to provide recommendations, aggregate statistics, or state-specific data? By providing results, we mean that each respondent has an opportunity after completing the survey to ask for us to send him or her a copy of the final report. Even if the respondent does not request the report, the final report will be published as a PATH publication and be will be available on the Internet and promoted through a HUD press release announcing new study results. Recommendations will be those included in the final report based on the analysis of the grouped data. Only aggregate statistics will be included in the final report. We may include some tables in the final report that are state specific, but none that are community specific.
Please describe how the frame for this study is developed. As a separate part of this study (secondary data analyses) we are looking at the regulatory codes of each CDBG-eligible community. We developed a coding scheme that establishes whether the community’s regulations fall into the low, mid, or high range based on a severity score. Those falling into the mid range are codes that are in a grey area and have some latitude for interpretation. Of those communities whose codes are determined to be in this grey area we planned to sample 250. To better our response, we will send a survey to all CDBG-eligible communities for which we can obtain address information rather than only sending a survey to a sample of communities having codes within the grey area of interpretation (250 is approximately the number of surveys that we anticipate receiving, given an anticipated low response rate. For example, if we are able to obtain addresses for 1,000 of the 1,839 jurisdictions, and we anticipate a 25% response rate, then we would get back 250 surveys. We base our expected return rate on our experience with previous projects.) Also, please elaborate on how CDBG-eligible communities will be grouped into the three categories? The grouping into three categories as identified through our devised coding scheme no longer applies to the mailed survey. We will send the survey to all communities for which we can obtain contact information. Please provide information on how the contact information for the respondents will be collected and how complete you expect this information to be. We are gathering contact information by searching the Internet for the CDBG-eligible community’s government website. We are searching for an address and the name of the Planning Director or of a person who would be familiar with local codes governing the placement of manufactured homes. When complete, we expect to have contact information for at least 1,000 communities. We will mail a questionnaire to all communities for which we have an address either using the using the contact name identified from our search or the generic term “Planning Director” if we cannot identify a specific contact name.
Please explain the statutory authority for informing respondents that their information is confidential. If the statutory authority doesn’t exist then respondents may be told that their information will be “protected.” If statutory authority exists, the law (or a citation) needs to be provided when assuring respondents of confidentiality. (This would be true on any materials provided to respondents.) We changed the wording from “information provided will be held in the strictest confidence” to “information provided will be protected”.
Please provide a copy of the cover letter and any additional mailing pieces that will be used in this study. Attached are cover letter first mailing, postcard reminder, and cover letter second mailing. See attachments manfhsgcoverletterfirstmailing.doc, manfhsgpostcard.doc, manfhsgcoverlettersecondmailing.doc. See attachment manfhsgquestionnaire.doc which is the questionnaire with the HUD ICR comment suggestions incorporated.
Please provide a copy of the script that will be used in the telephone interviews (for non-response follow-up). Telephone interviews are only offered in the second mailing as an alternative method of responding. We do not plan to telephone non-respondents. See attached telephone protocol. See attachment manfhsgphonecallprotocol.doc.
Regarding the mail operations, will all respondents receive a reminder postcard (and second mailing) or will a screening occur to only send these materials to the non-responders? Screening (based on an assigned ID control number printed on each survey) will occur to make sure that any respondent that has returned their survey will not receive a postcard or second mailing. How much time will there be between the first mailing and the subsequent mailings (reminder postcard, second mailing)? There will be two weeks between sending the first mailing and sending the reminder postcard and three weeks between sending the reminder postcard and sending the second mailing.
An anticipated response rate of below 80% is low. Have you considered the use of a pre-notice letter to potentially increase the response rate? We had not considered mailing a pre-notice letter (our mail procedure is based on a modified Dillman Method which does not suggest this). To increase the number of responses, we are mailing to all CDBG-eligible communities for which we can obtain contact information. While this may not increase the response rate, it should greatly increase our number of responses.
Is there an option in the first mailing for respondents to call in and complete a telephone interview? No. This may be a way to increase the response rate, particularly if you do have a call-in number for questions and are planning to do this during the second mailing? It could improve response rate, but we did not allow for this in the budget. We anticipate that by the second mailing there will be fewer respondents to whom we offer a telephone interview as a choice.
Since the mailing is going to an individual and that individual may choose to pass on the questionnaire for response, what measures will be used to ensure that follow-up mailings for non-response are targeted appropriately? If we receive a survey back whether completed by the original individual or another individual, no further communication will take place based on the ID number. However, if the original individual passes off the survey to someone else who does not respond within the time period between mailings, then the original respondent will receive the reminder postcard and/or second mailing. We assume the original respondent will pass on the reminders. If we receive a phone call that the questionnaire is being forwarded to another individual for completion, we will ask for the name of that individual and make the change on the master list. If we receive mail that provides the name of an individual other than the original respondent or generic “Planning Director”, we will make the change on the master list. These actions could help ensure follow-up mailings are targeted appropriately.
Please provide detailed information on the project schedule including information on the components of data collection, follow-up, and analysis.
Schedule for data collection, follow-up, and analysis – 12 months:
Preparation period – 8 months (some concurrent tasks)
Design, testing, approval of questionnaire – 3 months
Prepare cover letters for first and second mailing and text for postcard – 1 week
OMB approval – 4 months
Collect contact name and address information using the Internet to access government websites for up to 1,839 CDBG-eligible communities and enter that information into a spreadsheet- 4 months
Print questionnaires – 2 weeks
Print envelopes, both mail out and business reply return – 2 weeks
Create a master address file including ID control numbers from the spreadsheet of contact information – 2 days
Perform a mail merge based on the master list to create personalized cover letters on letterhead and mailing labels and print those letters and labels – 2 days
Create and print return address labels – 1 day
Apply return address labels to mail reply envelopes – 1 day
Apply ID control number to survey questionnaires – 1 day
Put together mail packets (applying respondent address label to envelope and enclose matching personalized cover letter, matching questionnaire, and return envelope) including accuracy check- 3 days
Mail out first mailing of questionnaire – 1 day
First mailing period – 2 weeks
Handle incoming mail from first mailing – 2 weeks
Designated researcher electronically records receipt of returned, completed questionnaires (entering a code of 1 on the master list indicates a completed return and prevents further mailings). A date of return is also recorded. If the respondent provides a change of address, the designated researcher will change the address on the master list.
File returned, completed questionnaires in secure location.
If the mailing is returned for insufficient address, we will check the community website. If the address on the returned mailing is correct according to the website, we will attempt to call the community to ask for an address. If we find an error was made or receive a new address based on a phone call, the designated researcher will correct the address on the master list and enter a code of 2 to indicate an address change and record a date for remailing. We will print a new label and remail the questionnaire. If we are unable to secure a new address, the designated researcher will enter a code of 3 on the master list preventing further mailings to that respondent.
If a respondent returns the survey but does not complete the survey, the designated researcher will enter a code of 4 on the master list preventing further mailings to that respondent.
Print reminder postcards – 2 weeks.
Preparation for reminder postcard – 2 days
Print respondent mailing labels based on updated master list (excludes respondents with codes of 1, 2, 3, or 4 (respondents coded 2 will be mailed a postcard approximately 3 weeks following remailing). – 1 day
Apply return address labels and respondent mailing labels to postcard – 1 day
Mail out reminder postcards – 1 day
Reminder postcard period – 3 weeks
Handle incoming mail continuing from first mailing and resulting from reminder postcard as outlined above under first mailing period– 3 weeks
Preparation for second mailing – 1 week
Perform a mail merge based on the updated master list to create personalized cover letters on letterhead and mailing labels and print those letters and labels (excludes respondents with codes of 1, 2, 3, or 4 (respondents coded 2 will be mailed a second letter and questionnaire approximately 3 weeks following the mailing of their postcard) – 2 days
Create and print return address labels – 1 day
Apply return address labels to mail reply envelopes – 1 day
Apply ID control number to survey questionnaires – 1 day
Put together mail packets (applying respondent address label to envelope and enclose matching personalized cover letter, matching questionnaire, and return envelope) including accuracy check. Pull of mailing packets for respondents from whom we received last minute completed questionnaires- 3 days
Mail out second mailing of questionnaire – 1 day
Post second mailing period – 6 weeks
Handle incoming mail continuing from first mailing and resulting from reminder postcard and second mailing as outlined above under first mailing period– 6 weeks
Accept or return phone calls from respondents requesting a telephone interview. Complete a blank questionnaire and record proper ID of caller. Enter code 5 on master list to indicate the questionnaire was obtained over the telephone.– 6 weeks
Send out second mailings delayed due to change of address – 1 day
Data Entry – 3 weeks
Create a master codebook to guide consistent data entry – 1 day
Enter data from completed questionnaires (including those completed from a telephone interview) into SPSS on a designated computer by a designated researcher. – 2 weeks
Check data – 1 week
Data Analysis – 2 weeks
Label variables in SPSS and create any new or manipulated variables – 2 days
Run frequencies on all variables and then run crosstabulations on selected variables – 2 weeks
Late returns (those received after the second letter and replacement questionnaire are mailed) are analyzed as a sample of non-respondents and compared with early returns. This comparison provides a test of possible bias related to differences between non-respondents and respondents. In addition, we will compare community characteristics (size, state, region) of respondents and non-respondents as an additional test of response bias. – 2 days
Are you intending to extrapolate the results of this study beyond the mid-range category of regulatory control? No longer applicable as results will be provided for all categories of regulatory control.
On Q2, is there a reason for the specific aggregations (i.e., 1-10, 11-20, … 51-99, …)? We offered grouped choices to save the respondent the time of looking up the exact number. There was not a basis for the specific aggregations that we choose other than they were reasonable, fairly standard groupings.
“DK” and “NA” need to be defined for the respondents.We made this change on the questionnaire.
| File Type | application/msword |
| File Title | Preliminary Comments on |
| Author | Lisa Blumerman |
| Last Modified By | Alex Hunt |
| File Modified | 2007-03-30 |
| File Created | 2007-03-30 |